NexRad Level II Data Format

1
NexRad Level II Data Format

• Reggie Mead
• 07/17/07

2
Background Information

• Nexrad radars measure certain physical attributes
  in the three dimensional space surrounding them.
• This three dimensional space is logically divided
  in a number of ways that mirror the way in which
  the data is physically collected. These logical
  distinctions also make it easier for data
  consumers to understand and manipulate the data
  in meaningful ways.
• The following abstractions are used to logically
  subdivide the data, each will be covered in the
  following slides
• Volume
• Sweep
• Scan
• Ray
• Pulse Volume

3
Volume

• A Volume represents the entire three dimensional
  space around a radar site at a single point in
  time (technically it is the span of time from
  when the radar starts collecting data to when it
  stops).
• A Volume is the top-most organizational
  construct for NexRad radar data.

4
Sweep

• Volumes are divided into sweeps.
• Each sweep has a unique elevation angle.
• Sweeps get their name from the operation of a
  NexRad radar. In order to scan a three
  dimensional space, the radar typically begins at
  the lowest elevation angle and rotates, sweeping
  out 360 degrees of space (one sweep).
• Next, the radar will raise its elevation angle
  and perform another sweep.
• This process is repeated until the entire three
  dimensional space has been scanned.

5
Scan

• Sweeps are themselves made up of a variable
  number of scans.
• Most of the time, a sweep does not collect all
  of the information it needs in a single 360
  degree rotation.
• Greater accuracy and precision can be achieved
  if each type of data is collected in its own
  rotation.
• Therefore, a sweep will often contain a number
  of scans at the same elevation angle.
• Scans can contain a single type of multiple
  types of data.

6
Ray

• Scans are subdivided into rays.
• Rays can be thought of as spokes on a tire.
• Typically, rays are about 1 wide, so scans
  contain roughly 360 rays (be aware that scans
  often rotate slightly more than 360 and will
  contain more than 360 rays).
• The angle of each ray is known as its azimuth.

7
Pulse Volume

• Each ray is divided into pulse volumes.
• Pulse volumes act like bins, catching the data
  for one block of two dimensional space.
• Pulse volumes typically have depths from 250m
  1000m in length.
• Pulse volumes are the atomic level of radar data
  and cannot be further subdivided.

8
Base Features

• NexRad radars collect three different base
  features
• Reflectivity
• Radial Velocity
• Spectrum Width
• Radial Velocity and Spectrum Width are known as
  Doppler data.

9
Reflectivity

• Reflectivity it is the traditional form of data
  captured by radars. It is a measure of the
  strength of the echo returned by a radar pulse.
  Reflectivity is measured in decibels and is
  usually denoted by the letter Z. Therefore,
  reflectivity values are often given units of dBz.

10
Radial Velocity Spectrum Width

• Radial Velocity measures the velocity of
  particles moving relative to the radar. Radial
  Velocity is measured in m/s.
• Spectrum Width is a measure of the variance of
  the doppler signals within a pulse volume.

11
Base Features

• NexRad radars collect three different base
  features
• Reflectivity
• Radial Velocity
• Spectrum Width
• Radial Velocity and Spectrum Width are known as
  Doppler data.

12
Level II Data Format

• Each file represents one Volume.
• Raw files have the .Z extension and are
  compressed using the LZC compression algorithm.

13
High Level Data Organization

• Files contain an initial header that is followed
  by a number of packets.
• Packets do not have to be the same size, but in
  practice they typically are.

14
Data Types

• Byte 8 bits, integer type
• Short 16 bits, integer type
• Int 32 bits, integer type
• Float 32 bits, floating type
• Unsigned Char 8 bits, integer type
• All data on disk is Big Endian

15
File Header Format
16
High Level Packet Organization
17
Message Header Information
18
Continued
19
Continued
20
Continued
21
Data Header Information
22
Continued
23
Pulse Volume Information
24
Data Parameters
25
Environmental Attributes
26
Data Frame Check Sequence
The array of data can be a mixture of
reflectivity, velocity, and spectrum width data.
If more than one type of data is present, the
data will always conform to the following
organization Reflectivity gt Velocity gt
Spectrum Width (i.e. if datax is reflectivity
data and datay is velocity data, then x lt y).
Values start at pulse volumes closest to radar
and get further from the radar as the index of
the data grows (i.e. datax is closer than
datax1). Bad values (below signal to noise
ratio) are given a value of 0 and range folded
values are given a value of 1).
27
Volume Coverage Patterns

• As mentioned earlier in this presentation,
  volumes contain a number of sweeps at different
  elevation angles and each sweep contains a number
  of scans.
• Volume Coverage Patterns define the number of
  sweeps, the elevation angles of those sweeps, as
  well as the number and types of scans in those
  sweeps.
• There are six common Volume Coverage Patterns
  that are used depending on atmospheric conditions
  and intended research.

28
Volume Coverage Pattern Categories

• Convection VCP11 and VCP12
• Shallow Precipitation VCP 21
• Clear Air VCP31, VCP32
• Multiple Pulse Frequency Dealiasing VCP 121

29
VCP 11

• Used for convection, especially when close to
  radar. Has the best overall volume coverage.
• 16 scans in 5 minutes

30
VCP 12

• Used for convection, especially at longer
  ranges. Focuses on lower elevations to better
  sample the lower levels of storms.
• 17 scans in 4 minutes

31
VCP 21

• Used for shallow precipitation. Rarely used for
  convection due to sparse elevation data and long
  scan time.
• 11 scans in 6 minutes

32
VCP 31

• Used for detecting subtle boundaries or wintry
  precipitation. Uses a Long pulse
• 7 scans in 10 minutes

33
VCP 32

• Slow rotation speed allows for increased
  sensitivity. This is the default clear air mode
  as it reduces wear on the antenna. Uses a short
  pulse
• 7 scans in 10 minutes

34
VCP 121

• Used when there are a large number of rotating
  storms, tropical storms, or when better velocity
  data is needed. Scans lower cuts multiple times
  with varying repetitions to greatly enhance
  velocity data.
• 20 scans in 5.5 minutes